Newly mapped brain circuit reveals how memories remain stable during learning

Summary: A new study reveals how brain circuits between the entorhinal cortex and hippocampus stabilize memories during learning. The researchers found that synchronized signals between excitatory and inhibitory neurons in these regions help strengthen “place maps,” ensuring that memories remain consistent even when new information is learned.

This balance of excitation and inhibition fine-tunes neural activity to filter important sensory information while preserving past knowledge. The findings offer new insights into how the brain encodes and maintains context-dependent memories and may inform treatments for disorders such as post-traumatic stress disorder and schizophrenia.

Key facts

Memory stability mechanism: Coordinated signaling between the entorhinal cortex and the CA3 region of the hippocampus stabilizes learned spatial maps. Dual neural action: Excitatory and inhibitory pathways work together to fine-tune activity and reinforce memory networks. Clinical relevance: Understanding these circuits may help develop therapies for the memory instability seen in post-traumatic stress disorder and schizophrenia.

Source: Langone of New York University

Newly decoded brain circuits make memories more stable as part of learning, according to a new study led by NYU Langone Health researchers.

Published online in Science on October 30, the study shows that activity in signaling pathways connecting two brain regions, the entorhinal cortex and the CA3 region of the hippocampus, helps mice encode place maps in brain circuits.

It is known from previous studies that the entorhinal/hippocampal circuit is crucial for both memory formation and memory retrieval by completing patterns from partial cues. Reliable memory requires that hippocampal place maps remain stable and resist changes in the environment to some extent.

Problems with CA3 neural calculations can lead to symptoms similar to those of schizophrenia or post-traumatic stress disorder, the study authors say, where the stability and precision of memories fail. In these cases, a balloon popping at a party could trigger a frozen fear response as a soldier’s brain mistakenly remembers a bomb exploding.

“Our study, by focusing on the stability of hippocampal representations, fills a substantial gap in the understanding of how long-range inputs control neural circuits essential for memory retrieval,” said the study’s senior author, Jayeeta Basu, PhD, an assistant professor in the departments of Psychiatry and Neuroscience at NYU Langone Health.

“A better understanding of the circuits that support place maps can guide the future design of more precise treatments for conditions that affect memory,” added Basu, a faculty member at the Translational Neuroscience Institute at NYU Langone Health and recent recipient of the Presidential Early Career Award for Scientists and Engineers.

Repeated Circuits Activity Set Memory Templates

The new study revolves around brain cells called neurons, which “fire” (or generate rapid changes in the balance of their positive and negative charges) to transmit electrical signals that coordinate thoughts and memories.

When a charge reaches the end of a brain cell’s extensions, it triggers the release of neurotransmitter chemicals that float across the space between one cell and the next. On the other hand, they are coupled to proteins that, depending on their nature, stimulate the activation of subsequent nerve cells (excitation) or inhibit their activation, say the researchers.

This combination of excitation and inhibition achieves a balance that sculpts “noise” in thoughts, a balance that is maintained when the brain is not learning (in a resting state). However, during learning, increases in arousal encode new memories, and the activity patterns of neurons determine the specificity of the memories they represent.

Reactivating these neurons in a set pattern recalls a specific memory and produces the related behavior, such as a mouse learning where the sugar water rewards are in one maze versus another.

The current study focuses on neurons with long extensions that coordinate activity between distant brain regions. Little is known about how long-range cell inputs influence local circuits as the brain balances stable templates (of what is already known) with new data (about constantly changing experiences) to form memories.

The research team determined that two types of long-range extensions from the lateral entorhinal cortex to the CA3 region send signals at the same time to stabilize the activity of the brain cells’ learning networks. Specifically, long-range excitatory glutamatergic (LECGLU) and inhibitory GABAergic (LECGABA) extensions were found to increase the activity of sets of interconnected neurons to support learning.

The study authors examined interactions between long-range LEC inputs and CA3 circuits at the single-cell level. LECGLU was found to drive excitation in CA3 but also feed-forward inhibition that fine-tunes firing, while LECGABA suppresses this local inhibition to disinhibit (promote) CA3 activity. This combined action supported stability in CA3 by triggering recurrent activity in certain circuits, encoding place memories.

“This work looked at the mechanism by which the brain increases the excitation of brain cells to pay more attention to certain sensory information by reducing inhibition in key microcircuits,” says study first author Vincent Robert, PhD, a postdoctoral researcher in Basu’s lab.

“The team detailed a circuit mechanism that fine-tunes the cross-talk between excitation, inhibition, and disinhibition in the service of context-dependent memory formation and place map stability.”

Along with Basu and Robert, study authors from the Department of Neuroscience at NYU Langone Health are Keelin O’Neil, Jason Moore, Shannon Rashid, Cara Johnson and Rodrigo De La Torre of the Basu lab. Other authors are Boris Zemelman of the Center for Learning and Memory at the University of Texas, Austin, and Claudia Clopath of the Department of Bioengineering at Imperial College London, Basu’s co-principal investigator for an R01 grant from the National Institutes of Health (NIH) BRAIN Initiative.

Funding: Funding for the study was provided by NIH grants 1R01NS109994, 1R01NS109362-01, 1RM1NS132981-01, 5T32MH019524-30, training grant T32GM007308, 3R01MH122391-04S1, R01MH122391, 1U01 NS099720 (BVZ) and 1U01 NS094330. Also providing support were a McKnight Scholar Award in Neuroscience, the Klingenstein-Simons Fellowship Award in Neuroscience, an Alzheimer’s Association Research Grant to Promote Diversity, a Sloan Research Grant, a Mathers Foundation Award, a Whitehall Foundation Research Grant, an Epilepsy Society of America Junior Investigator Research Award, a Blas Frangione Young Investigator Research Grant from New York University, a Leon Levy Foundation Award, a Bettencourt Award for Young Researchers and the Emerald Foundation.

Key questions answered:

About this research news on learning and memory.

Author: Gregory Williams
Source: Langone of New York University
Contact: Gregory Williams – NYU Langone
Image: Image is credited to Neuroscience News.

Original Research: Closed access.
“Glutamatergic and GABAergic cortical inputs support learning-driven hippocampal stability” by Jayeeta Basu et al. Science

Abstract

Cortical glutamatergic and GABAergic inputs support learning-driven hippocampal stability

The flexibility and stability of neuronal ensembles are crucial characteristics of brain function. Little is known about how these local circuit properties are influenced by long-range inputs.

We show that glutamatergic (LECGLU) and GABAergic (LECGABA) projections from the entorhinal cortex lateral to CA3 recruit specific microcircuits that jointly provide stability to the neuronal ensembles that support learning. LECGLU drives excitation in CA3 but also substantial inhibition of feeding that prevents somatic and dendritic spikes.

In contrast, LECGABA suppresses this local inhibition to disinhibit CA3 activity with compartment and pathway specificity by selectively increasing somatic output to the integrated recurrent inputs of LECGLU and CA3.

This synergy allows the stabilization of spatial representations relevant to learning, as both LECGLU and LECGABA control the formation and maintenance of CA3 place cells across contexts and over time.